SUMMARY Macrolide antibiotics are among the most clinically important antibacterial agents. They inhibit cell growth by binding to the ribosome in the nascent peptide exit tunnel and interfering with protein synthesis. Crystallographic structures of the ribosome complexed to macrolide antibiotics revealed the main sites of contact, including the key interaction with adenine residue A2058 (using Escherichia coli numbering), which is universally conserved in bacteria. One of the main mechanisms of resistance to macrolides is dimethylation of A2058 by the Erm-type rRNA methyltransferases. This modification also confers resistance to other classes of ribosome targeting antibiotics, including lincosamides and streptogramins B. The structure of the macrolide- binding site in the Erm-modified ribosome is unknown. Although possible explanations for the effect of A2058 dimethylation on drug binding have been proposed, they cannot account for the available experimental data. Therefore, at the moment, there is no satisfactory explanation of why dimethylation of A2058 by Erm methyltransferase prevents antibiotic binding. The lack of this knowledge precludes rational development of newer macrolides and other antibiotics that would maintain clinically relevant activities against the Erm- modified ribosome. In the proposed project, we will solve this long-standing problem by obtaining the crystallographic structure of the Erm-modified bacterial ribosome in isolation or in complex with antibiotics exhibiting residual activity against the A2058 dimethylated ribosome. To achieve this goal, we will optimize the expression of functionally- active Erm methyltransferases in the cells of thermophilic bacterium Thermus thermophilus, which has been used as the source of ribosomes suitable for crystallographic studies. We have found that in spite of growing optimally at 72-75C, T. thermophilus can grow at notably lower temperatures. By cloning and expressing Erm- methyltransferase genes from moderately thermophilic bacteria in T. thermophilus grown at high temperatures or by expressing the erm genes from the mesophilic bacteria in the T. thermophilus host grown at reduced temperatures, we will obtain the crystallizable ribosome dimethylated at A2058. Once the atomic structure of the vacant Erm-modified ribosome is obtained, we will solve the structures of the Erm-modified ribosomes complexed with several macrolides which retain residual activity against the Erm-modified ribosome. The resulting information will be instrumental for understanding the molecular principles of Erm-mediated resistance and will inform the subsequent rational design of new antibacterials active against Erm-positive pathogens.